jquazza

When a document says the adp/acp tests will be tested according to the regulations, that is incorporating the statutory exclusions rules by reference. And by the way, I have seen attorney drafted documents where the exact method of testing is described without mention of the regs. Which means when there are changes to the regs, the attorney has to amend the plan to conform and gets more fees...

I have to deal with some of these auditors as well. Basically, you should point them to the Schedule H instructions (line 1h.) It specifies that noncash basis plans should include distributions only if they were approved as of the last day of the PY but had not been paid. How were these corrective distributions approved if the tests had not even been completed?

Out of curiosity, how many of these late filers were DRs?

I agree with WDIK. This option is used by certain financial institutions (Banks, Insurance Companies etc...) who sponsor CCTs, PSAs, MTIA etc... If a 5500 is filed for the CCT or PSA..., then when you complete shchedule H for a plan that invested in such contract, you do not have to report the underlying assets, you only have to report that the plan invested so much in PSA etc... If your plan had more than 100 participants at the beginning of the year, you have to file a schedule H and get an accountant's opinion.

You always have to follow the terms of your plan document. If your document was written in such a way that it does not give you any flexibility on the method used for testing, you have to follow the terms of your document. Thankfully, most documents aren't drafted like that.

They have to be incorporated by reference. If your document specifies the manner in which the plan will be tested without referencing the code sections and regulations, then you have to follow the terms of your document.

I feel vindicated, thank you WDIK!

Okay, I’ll illustrate with an example: Assume 20k in an IRA, rate of return 10%, tax bracket 28% for 5 years: IRA: After 5 years, balance is $32,210.20 –$9,018.87 (taxes) = $23,191.33 Invest the $5,600 would pay in taxes on the Roth at same rate of return and tax bracket = $7,927.97 Total: $31,119.30 Roth IRA: $32,210.20 tax free My point is there is a difference, which would only get larger as time and earnings increase.

You don't have to pay the taxes with the IRA assets. Take a $20k distribution from an IRA, roll 20k into a roth, pay the taxes with your other savings... watch it grow tax free. Is your answer still the same?

As far as the 15% penalty is concerned, we usually use the underpayment penalty rate for the applicable period, however, this should truly be calculated based on how the sponsor benefited. For example, the sponsor has a line of credit at the bank which would have been used to make the deposit. The prohibited transaction resulting from the use of the plan assets benefited the sponsor because he didn't have to tap in his line of credit to make the deposit, it would therefore be reasonable to use the line of credit interest rate as the base of earnings for the penalty calculation. The same logic would apply if the sponsor hold his funds in an account at a bank that's earning a specified interest. For all you know, the sponsor might have used the money to buy Lotto tickets and hit the jackpot, that how he benefited with the plan assets and that's what the penalty calculation should be based on. For the make up earnings, there are several methods acceptable, actual rate of return of the plan investments (w/o regards to losses) best performing fund or the underpayment penalty rate (published quarterly in an IRS ruling, the latest one is 2004-56.)

I beg to differ on your math. Investment earnings in a Roth IRA are tax free whereas on a 403(b) or Traditional IRA, they are only tax deferred, so, in a Roth, you will pay 0% taxes on the earnings while on the other vehicles, you will pay taxes on the earnings based on your tax bracket. If you stay in the same tax bracket, you therefore have to consider the potential earnings of the taxes you paid upfront in a Roth IRA versus the tax free earning potential. If you're not going to need the money until retirement, considering your age, say in 25-30 years, chances are you would be better in a Roth.

yes, provided the plan was amended for egtrra

Do you even have any contributions and/or compensation during that period? If you don't, you don't really have an issue, if you do (i.e. 12/31 was a payroll date), it would seem reasonable to prorate the limits you use in your tests to the payroll period covered (i.e. monthly: 1/12, biweekly: 1/26.) The 401(a)(17) regs actually mention the proration to the number of months in the plan year, but your case is so extreme (a 1-day plan year) that you may have a hard time justifying that 1 day = 1 month.

To kick it up another notch, on Schedule T, I would report two exceptions: -Box 3D for the 401(k) (providing there were no exclusions.) -Box 3B for the other components of the plan (as no HCEs benefited in these sources.)

I thought Partners who wanted to make deferral contributions up to the tax return deadline of the partnership could, however, they needed to make a formal written election before the end of the plan year (which could be contingent on the partners'income.)

TBOB, MRDs can be subject to the 20% W/H rule in the case of non 5% owner who are still employed and decided to take the MRDs, since it's an MRD that's not required so to speak.

Blinky, it's true, I have seen a lot of plans lately, that have received a FDL from the IRS, and they had each partner, identified by name, in its own class. Did the IRS pull a switcharoo? What happened? How will that affect your reasonable classification test? Also, in previous post, it was mentioned that if an eligible class gets a 0 contribution rate (and others are getting something), the plan was restricted from using the ABT. Is the IRS taking names and these plans will be the first ones to be audited? I mean, how good can a new comp be if you can't use ABT?

I never really understood the sponsor who want to avoid a failing test by all means, to me if you're plan is not failing, that means (most of the time) that the HCEs aren't deferring as much as they could. The argument about the discretionary match on a prior year method though valid could be easily set aside by selecting the prior year method for ADP and current for ACP.

Which brings you to the correction methods. If the sponsor did not withhold deferrals from bonuses and the plan doesn't exclude bonuses from comp nor does it provide for a separate election, the sponsor will probably have to make a QNEC to make up for the missed deferrals.

One more thing, about the rounding, you always have to round up. if your 20% count is 4.1, you have to include 5 individuals as your HCEs.

When was the last time you've seen a plan administrator run a credit report on a participant?

I think that's arguable. The basic principle of EPCRS is to make the participants whole as if the error had never occured, HCEs or not.

I would use the rule that excludes the employees w/ less than six months of service, under 21, who do not work at least 17.5 hours/week etc... Keep in mind you can only exclude the CBA employees if they don't benefit under the plan.

Assuming the test is now correct and cannot be tweaked to get to the amounts that were refunded, obtaining the repayment from the participant may not be an easy thing to do. I believe the plan administrator should make these participants whole, even if they don't returned the money, as he approved the impermissible distribution.

I was also under the impression that designating groups by name was not acceptable, however, I have seen a lot (maybe a dozen) of plan documents that did just that and that were approved by the IRS, which seems to support k-man position. There are few issues you have to be very wary about, one of them, mentioned by QDROphile (maybe with a little lack of tact,) is that the fact the contribution could be construed as disguised CODAs. There is also the issue of not being able to run the ABT if some groups have a zero contribution rate, which in most cases, would defeat the purpose of setting up rate groups.